We have formulated a developmental neuro-cognitive model of lexical processing in normal readers based on cross-sectional functional magnetic resonance imaging (fMRI) data and we have collected our first longitudinal data on these children 2 years after their initial assessment. The proposed grant will longitudinally follow normal children (7- to 17- year-olds) and children with reading disorders (11- to 15-year-olds) when they perform a variety of tasks. These tasks include orthographic (spelling), phonological (rhyming) and semantic (meaning) judgments in the visual and auditory modalities. Not only is this proposal innovative because we use a variety of tasks, but we also have multiple perceptual controls and multiple parametric manipulations of difficulty, so we can more effectively examine developmental and group effects specific to aspects of lexical processing. Furthermore, our paradigms allow an examination of priming and lexical effects (e.g. word frequency, phonological consistency and orthographic consistency) so observed developmental and group differences are not likely to be due to specific characteristics of the tasks. The overall aim of this study is to determine whether different groups have different developmental trajectories. A longitudinal design is crucial to rule out cohort effects and to examine the predictability of the individual's subsequent brain development from their earlier reading skill and brain activation patterns. In addition to examining whether higher and lower skill normal readers have different developmental trajectories, we will examine whether dyslexic readers with specific deficits in decoding orthographic stimuli have different developmental trajectories from language-impaired readers with deficits in decoding orthographic stimuli in addition to general language processing deficits. We will also compare children with disorders to age-match versus reading-match normal readers to address the question of developmental delay versus deviance. Our general hypothesis is that individual differences in brain activation between readers will increase with age. This will be tested using hierarchical linear modeling to examine growth curves in the rate of change (e.g. slope) and in the shape of change (e.g. acceleration). We will use this technique to look at developmental changes in signal intensity in our critical regions of interest, but also for changes in effective connectivity using Dynamic Causal Modeling (DCM). PUBLIC HEALTH RELEVANCE This proposal will give us basic information about how the brains of typical children change throughout childhood and adolescence during reading and language processing, but also how the brains of children with reading and language disorders differ in their development. The relevance of this grant to biomedical issues is that it has implications for diagnosis and intervention in children with reading and language disorders.